There's an ongoing discussion about this. It seems it has to do with the
hitless ARQ modulation change support.

Basically, it's a trade-off.

On Jan 13, 2017 10:14 PM, "Mike Hammett" <af...@ics-il.net> wrote:

> I love how Gary says they use so much better, newer technology...  but
> they get worse throughput.  *sigh*
>
>
>
> -----
> Mike Hammett
> Intelligent Computing Solutions <http://www.ics-il.com/>
> <https://www.facebook.com/ICSIL>
> <https://plus.google.com/+IntelligentComputingSolutionsDeKalb>
> <https://www.linkedin.com/company/intelligent-computing-solutions>
> <https://twitter.com/ICSIL>
> Midwest Internet Exchange <http://www.midwest-ix.com/>
> <https://www.facebook.com/mdwestix>
> <https://www.linkedin.com/company/midwest-internet-exchange>
> <https://twitter.com/mdwestix>
> The Brothers WISP <http://www.thebrotherswisp.com/>
> <https://www.facebook.com/thebrotherswisp>
>
>
> <https://www.youtube.com/channel/UCXSdfxQv7SpoRQYNyLwntZg>
> ------------------------------
> *From: *"Tim Hardy" <tha...@comsearch.com>
> *To: *af@afmug.com
> *Sent: *Friday, January 13, 2017 5:54:53 PM
> *Subject: *Re: [AFMUG] Our UBNT AF11x experience
>
> Post 6 here discusses the use of ARQ rather than FEC.
>
> https://community.ubnt.com/t5/airFiber/AF11x-Capacity/td-p/1737631
> ------------------------------
> *From:* Af <af-boun...@afmug.com> on behalf of George Skorup <
> geo...@cbcast.com>
> *Sent:* Friday, January 13, 2017 6:41:10 PM
> *To:* af@afmug.com
> *Subject:* Re: [AFMUG] Our UBNT AF11x experience
>
> Which I don't understand. Sounds like strong FEC?
>
> On 1/13/2017 5:36 PM, Ken Hohhof wrote:
>
> According to Link Planner, PTP820S (IP20S) gets around 243M at 256QAM,
> 347M at 2048QAM (350M with header compression).  So add that to George’s
> Trango/Exalt/SAF list.
>
>
>
> *From:* Af [mailto:af-boun...@afmug.com <af-boun...@afmug.com>] *On
> Behalf Of *Chris Gustaf
> *Sent:* Friday, January 13, 2017 5:20 PM
> *To:* af@afmug.com
> *Subject:* Re: [AFMUG] Our UBNT AF11x experience
>
>
>
> Real usable capacity at Layer 2 Ethernet depends on several variables
> besides the modulation level and the regulatory bandwidth.  For 1024 QAM
> there are 10 bits per symbol so theoretically one would expect to get 400
> Mbps in a 40 MHz channel (10 bits x 40 MHz symbol rate).
>
>
>
> To actually make a radio that meets FCC spectrum masks, a reduction of the
> symbol rate down to around 35 MHz is required plus filtering, giving a max
> of 350 Mbps over the air.
>
>
>
> After that, Forward Error Correction (FEC) is required which is typically
> around 85-90 % for most microwave radios using LDPC, giving a net
> throughput of around 300 to 315 Mbps.
>
>
>
> Add header compression at layer 2 which in the worst case for large
> packets adds about 2.5 % improvement and you get to around 325 Mbps, which
> is what Trango and others get on a single 40 MHz channel with the newest
> technology.
>
>
>
> Ubiquiti maybe using a lower FEC rate like 75-80% (or a lower symbol rate
> if the filtering is not steep enough) and no header compression, which
> would give them about 250 Mbps for large packets.
>
>
>
> Hope that helps!
>
>
>
> Chris Gustaf
>
> Trango Systems
>
>
>
> On Fri, Jan 13, 2017 at 2:39 PM, Mathew Howard <mhoward...@gmail.com>
> wrote:
>
> What other radios will only do 250Mbps (502.4Mbps capacity according to
> the data sheet...) on one 40mhz single polarity channel at 256qam? The
> other's I've looked at are typically around 300Mbps...
>
> Not that I'm saying the AF11 is a bad radio, I plan on putting up at least
> a few of them this year... but I find it a little funny that they use
> 1024qam as a selling point, when it can't do as much throughput at 1024qam
> as other radios do at 256qam.
>
>
>
> On Fri, Jan 13, 2017 at 4:19 PM, Eric Kuhnke <eric.kuh...@gmail.com>
> wrote:
>
> I think this whole thread can be summed up that there's no magical way to
> get around the shannon limit and bps/Hz coding efficiency for a given
> channel size and modulation...   From the technical perspective of people
> who have to really understand the FDX capacity of a new PTP link, the ubnt
> marketing department is divorced from reality, it's not 1200 Mbps no matter
> how many slick PDFs they publish.
>
> One 40 MHz channel single polarity at 1024QAM 5/6 code rate is going to be
> nearly the same efficiency bps/Hz from many different manufacturers.
> Multiply as necessary for dual polarity or for larger channel sizes such as
> 60, 80 or 112 MHz.
>
>
>
> On Fri, Jan 13, 2017 at 9:38 AM, Faisal Imtiaz <fai...@snappytelecom.net>
> wrote:
>
> FYI... on a "single"  40mhz channel 'dual polarity' (mimo) the expected
> throughput on Af11x is 500meg duplex.
>
>
>
> Faisal Imtiaz
> Snappy Internet & Telecom
> 7266 SW 48 Street
> Miami, FL 33155
> Tel: 305 663 5518 x 232 <%28305%29%20663-5518>
>
> Help-desk: (305)663-5518 <%28305%29%20663-5518> Option 2 or Email:
> supp...@snappytelecom.net
>
>
> ------------------------------
>
> *From: *"Josh Baird" <joshba...@gmail.com>
> *To: *af@afmug.com
> *Sent: *Friday, January 13, 2017 10:30:10 AM
>
>
> *Subject: *Re: [AFMUG] Our UBNT AF11x experience
>
> But if I can get 80mhz channels in both polarities (running at
> 56Mhz/1024QAM with this radio), I should be able to at least double he
> capacity of my PTP-800 link which can do 228Mbps.  Right?
>
>
>
> On Thu, Jan 12, 2017 at 5:55 PM, Mike Hammett <af...@ics-il.net> wrote:
>
> At 40 MHz and a single polarity, you're looking at an almost insignificant
> increase in throughput.
>
> Their claim is 1.2 gb+.
> Cut that in half as they're advertising the aggregate, so 600 mb+.
> That's using both polarities, so now only 300 mb+.
> Only I haven't heard of anyone getting much more than 500 in a single
> direction (they may certainly exist, I just haven't seen them), so now that
> 300 is really only 250.
>
>
> Not much of an upgrade unless you can also get larger channels in both
> polarities.
>
>
>
> -----
> Mike Hammett
> Intelligent Computing Solutions <http://www.ics-il.com/>
>
> *Midwest Internet Exchange*
>
> *The Brothers WISP*
>
>
>
> <https://www.facebook.com/ICSIL>
> ------------------------------
> <https://www.facebook.com/ICSIL>
>
> *From: *"Josh Baird" <*joshba...@gmail.com*>
> *To: **af@afmug.com*
> *Sent: *Thursday, January 12, 2017 11:47:42 AM
> *Subject: *Re: [AFMUG] Our UBNT AF11x experience
> <https://www.facebook.com/ICSIL>
>
> This PTP800 is only capable of running at 40Mhz (ODU-A) so it can only do
> 228Mbps full-duplex.  The AF11x should be able to do much more than that,
> right? <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> On Thu, Jan 12, 2017 at 12:05 PM, Gino Villarini <*g...@aeronetpr.com*>
> wrote: <https://www.facebook.com/ICSIL>
>
> IIRC, PTP800 is Remec Style, you'll need Remec to N connector adapters
> (AF11x is N) <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> What do you expect to achieve with this upgrade?  Not much capacity
>  difference between PTP800 and AF11x, maybe 50-80- mbps more.  Only if you
> have a xpic license you can double your throughput with the af11x
> <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> *From: *Af <*af-boun...@afmug.com*> on behalf of Josh Baird <
> *joshba...@gmail.com*>
> *Reply-To: *"*af@afmug.com*" <*af@afmug.com*>
> *Date: *Thursday, January 12, 2017 at 9:12 AM
> *To: *"*af@afmug.com*" <*af@afmug.com*>
> *Subject: *Re: [AFMUG] Our UBNT AF11x experience
> <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> Good news - thanks for sharing. <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> Somewhat un-related question: <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> I have a PTP-800 link using these dishes: <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>
> *http://www.hol4g.com/AC/product.aspx?number=ANC-VHLP3-11W-RR1&p=237127&sc=0*
> <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> Do you know if I can re-use these dishes with the AF11x?  Do I need
> adapters? <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> Josh <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>
>
> *Gino Villarini*
>
> President
>
> Metro Office Park #18 Suite 304 Guaynabo, Puerto Rico 00968
>
>   <https://www.facebook.com/ICSIL>
>
> <https://www.facebook.com/ICSIL>
>
> On Thu, Jan 12, 2017 at 8:07 AM, Gino Villarini <*g...@aeronetpr.com*>
> wrote: <https://www.facebook.com/ICSIL>
>
> Hey all, just dropping by to share our experience with AF11x, we habe been
> beta testing the unit since Sept and for the last 3 months, the unit has
> been rock solid.    <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> We are very happy with its performance, just wished it had a SFP port!
> <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> This unit replaced a Mimosa B11 unit that we were having some intermitent
> throughtput issues,  <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> The swap was easy since we reused the Jirous Dishes and only had to add
> the af11x adapters to it,  <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> The link  went live on 9/21/16 and on the first weeks we experienced some
> lockups, but after a revised beta fw was applied, all issues went away.
> <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
> For UBNT, please add SFP port and continue the good work towards a af6x
> and af18x <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>
>
> *Gino Villarini*
>
> President
>
> Metro Office Park #18 Suite 304 Guaynabo, Puerto Rico 00968
>
>   <https://www.facebook.com/ICSIL>
>
> <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>   <https://www.facebook.com/ICSIL>
>
>
>
>

Reply via email to